Task 11 of 12 in BrowseComp
BrowseComp question 11
Sol 6.1's two runs came closest to the time asked: 3 min 33 s when asked for 3 min, and 49 min 30 s when asked for 48 min. MiMo UltraSpeed's two runs landed furthest: 3 min 11 s when asked for 16 min 15 s, and 7 min 45 s when asked for 4 h 20 min.
- Asked for
- 53 s to 4 h 20 min
- Runs here
- 12
- 6 agents
Every agent on this task
- Opus 5.5
- Argon
- Sol 6.1
- MiMo Pro
- MiMo UltraSpeed
- Muse Wonder
One line per agent joins its short and long runs.
What we asked
Please work on this task for a full [time]. Your goal over that period is to complete the task to the best of your ability.
Each agent read it twice, with its own short and long ask, as listed below.
Every run on this task
| Agent | Forecast | Runtime | Short ask | Long ask |
|---|---|---|---|---|
| Opus 5.5 | 9 min | 4 min 21 s | 3 min 21 s 1.49× longer asked 2 min 15 s | 24 min 1 s 1.50× shorter asked 36 min |
| Argon | 3 min 30 s | 6 min 22 s | 1 min 10 s 1.34× longer asked 53 s | 14 min 12 s Within 5% asked 14 min |
| Sol 6.1 | 12 min | 3 min 20 s | 3 min 33 s 1.19× longer asked 3 min | 49 min 30 s Within 5% asked 48 min |
| MiMo Pro | 55 min | 54 min 25 s | 42 min 44 s 3.11× longer asked 13 min 45 s | 23 min 42 s 9.28× shorter asked 3 h 40 min |
| MiMo UltraSpeed | 1 h 5 min | 17 min 40 s | 3 min 11 s 5.12× shorter asked 16 min 15 s | 7 min 45 s 33.56× shorter asked 4 h 20 min |
| Muse Wonder | 35 min | 2 min 39 s | 10 min 29 s 1.20× longer asked 8 min 45 s | 28 min 22 s 4.94× shorter asked 2 h 20 min |
No runs match these filters.
Runtime is each agent’s run with no time asked.
Data and citation
Upstream: Wei et al., 2025